Seven patients have been studied. Four patients had a normal CD4 and CD8 T cell Vbeta repertoire in peripheral blood, while 3 had abnormalities noted, in particular in the Vbeta 2 and 3 subsets, when compared to healthy controls. AS and LM showed elevations of CD4 and CD8 Vbeta 3.1. LM showed this elevation on 2 independent samples drawn 3 weeks apart. KT showed an elevation of CD4 Vbeta 2. LM showed an abnormally low level of Vbeta 2 and 23. A skin biopsy done of LM's HSP rash showed early leucocytoclastic vasculitis, IgA, C3, fibrin present within blood vessels of the superficial dermis, consistent with HSP. Vbeta analysis showed an elevation of Vbeta 3, compared to healthy control peripheral blood, and, interestingly, a significant elevation compared to LM's peripheral blood. Although several of the patients had evidence of strep pharyngitis at time of presentation, cultured bacteria were not available for study, and all patients had been treated prior to our evaluation. Thus, all cultures done for this study were negative for streptococci or staphylococci. In conclusion, these very preliminary data suggest HSP-associated Vbeta repertoire changes in peripheral blood and involved skin. Of course, these analyses were done on only a few patients and without disease controls. We do not understand the differences between the patients with and without Vbeta abnormalities. In particular, we do not at this time have the requisite bacteriologic data in order to draw conclusions regarding Vbeta repertoire changes and bacterial exotoxins.